Anjal Parikh

Andrej Karpathy just joined Anthropic! Seems like they are going to be kicking things up a notch with future releases 👀

@AnthropicAI you guys should really let me interview for DevRel! We’ve done tons of free workshops and events helping people use Claude and its resources! Day #1 of trying to get noticed and picked up by @AnthropicAI @claudeai

@ClawPlexDFW @25NCoworking @Amit_0717 We cooked with this event. Insane job @Amit_0717 and thank you so much for letting me present!

Huge thank you to everyone who showed up to the Claude in the Wild event today! It feels great to be in the room with others who are as enthusiastic about AI technologies as we are. Special shout-out to @25NCoworking for giving us the space! @Amit_0717 is a legendary host!

@DataChaz I deployed Hermes on Macmini too and it’s flawless. I just like having it setup in its own workspace.

OpenClaw users who bought a Mac Mini right now:

Spreading the word on @NousResearch Hermes Agent — the self-improving AI agent that grows with you. Join us June 10th at Spark Coworking in Arlington, TX for a deep dive into this open-source autonomous agent from Nous Research. 🗓 Wed, June 10 | 4–6 PM 📍 @SparkArlington, Arlington, TX

Here’s my server setup ao I can pull context from my @obsdmd vault based on the channel I am in for both my agents to review before they answer!

I’ve been in love with @openclaw for quite some time. A die hard fan to be precise. Most recently my friend @Amit_0717 showed be @NousResearch ‘s Hermes tool and boy I haven’t looked back. I spent 30 hours deep diving and setting everything up from scratch with Calliope!

3. The self learning feature helps calliope retain and differentiate what done looks like, previous comments and concerns and how to make sure the output is concrete, backed with proof or work to show.

2. Calliope builds her own skills knows how I like things done and has been a great resource overall. She and Athena, my openclaw execute tasks in parallel where Calliope helps guide my claw through a harness I built.

1. Calliope sits on my discord and has a ton of channels! Everything is broken down based on my goals, dreams, and where I want to get to in the next 5 years.

@Cap_N_Chronic @JasperMSMKT you’re affiliated with boardy? I’ve used him a ton. Pretty cool!

@Cap_N_Chronic luma.com/u3e9qs8i luma.com/aitx-dfw-may26 Incase yall wanna join!

@Cap_N_Chronic Need permissions to talk! Sup guys!

This is going to be game changing for power users. No more 8 terminal sessions with Claude code running on all of them! @AnthropicAI just cooked!

If you are in DFW you should definitely come by! Good times, great networking opportunities and a really good speaker lineup.

I spent 10-15 minutes playing on the new Codex and I have to say I am mind blown! Big attribution to the devs @OpenAI @steipete ! Very good stuff coming out of OpenAI!

@steipete @NoahHirshon I’m rooting for you to make GPT better. I don’t understand what’s happening at Anthropic anymore. It’s sad….

@NoahHirshon I'm biased, but just making GPT really good character-wise.

Anthropic now blocks first-party harness use too 👀 claude -p --append-system-prompt 'A personal assistant running inside OpenClaw.' 'is clawd here?' → 400 Third-party apps now draw from your extra usage, not your plan limits. So yeah: bring your own coin 🪙🦞

LLM Knowledge Bases Something I'm finding very useful recently: using LLMs to build personal knowledge bases for various topics of research interest. In this way, a large fraction of my recent token throughput is going less into manipulating code, and more into manipulating knowledge (stored as markdown and images). The latest LLMs are quite good at it. So: Data ingest: I index source documents (articles, papers, repos, datasets, images, etc.) into a raw/ directory, then I use an LLM to incrementally "compile" a wiki, which is just a collection of .md files in a directory structure. The wiki includes summaries of all the data in raw/, backlinks, and then it categorizes data into concepts, writes articles for them, and links them all. To convert web articles into .md files I like to use the Obsidian Web Clipper extension, and then I also use a hotkey to download all the related images to local so that my LLM can easily reference them. IDE: I use Obsidian as the IDE "frontend" where I can view the raw data, the the compiled wiki, and the derived visualizations. Important to note that the LLM writes and maintains all of the data of the wiki, I rarely touch it directly. I've played with a few Obsidian plugins to render and view data in other ways (e.g. Marp for slides). Q&A: Where things get interesting is that once your wiki is big enough (e.g. mine on some recent research is ~100 articles and ~400K words), you can ask your LLM agent all kinds of complex questions against the wiki, and it will go off, research the answers, etc. I thought I had to reach for fancy RAG, but the LLM has been pretty good about auto-maintaining index files and brief summaries of all the documents and it reads all the important related data fairly easily at this ~small scale. Output: Instead of getting answers in text/terminal, I like to have it render markdown files for me, or slide shows (Marp format), or matplotlib images, all of which I then view again in Obsidian. You can imagine many other visual output formats depending on the query. Often, I end up "filing" the outputs back into the wiki to enhance it for further queries. So my own explorations and queries always "add up" in the knowledge base. Linting: I've run some LLM "health checks" over the wiki to e.g. find inconsistent data, impute missing data (with web searchers), find interesting connections for new article candidates, etc., to incrementally clean up the wiki and enhance its overall data integrity. The LLMs are quite good at suggesting further questions to ask and look into. Extra tools: I find myself developing additional tools to process the data, e.g. I vibe coded a small and naive search engine over the wiki, which I both use directly (in a web ui), but more often I want to hand it off to an LLM via CLI as a tool for larger queries. Further explorations: As the repo grows, the natural desire is to also think about synthetic data generation + finetuning to have your LLM "know" the data in its weights instead of just context windows. TLDR: raw data from a given number of sources is collected, then compiled by an LLM into a .md wiki, then operated on by various CLIs by the LLM to do Q&A and to incrementally enhance the wiki, and all of it viewable in Obsidian. You rarely ever write or edit the wiki manually, it's the domain of the LLM. I think there is room here for an incredible new product instead of a hacky collection of scripts.

@martinamps @3rosika @Prince_Canuma @AnthropicAI Hey Martin! I have a Claude dev and partner question. We use Claude heavily and I’ve sent out over 40 referrals to people who join our startup. Is there a way we can apply to become a partner? I’d love to be an advocate if possible!

@3rosika @Prince_Canuma @AnthropicAI dm me your email and i’ll check!

Hey @AnthropicAI I was accepted into your OSS program with 20x free usage, but my card is still getting charged for the Max plan. Support keeps routing me to a bot that can’t resolve this. Is there a way to reach an actual person?

@cryptopunk7213 I don’t think it killed the industry. Not all feature updates decimate industries overnight. Will people get dropped by this feature? Probably. You’d be better off using the feature with PR & Vulnerability reviews before pushing to prod. Will it make it faster? 100%

this is fucking ridiculous lol - anthropic just killed a $50B industry with a single feature (again): - companies pay $50K a year to scan their code for vulnerabilities. - anthropics Code Review does it for you in minutes for a fraction of the cost. - deploys multiple agents to hunt for bugs in your code. internal results show its amazing (84% hit rate on 1000+ line code base) for comparison: anthropic cost = $15-25 PER review, trad competitor cost = $99+ complete fucking no brainer. watch the appsec stocks react to this one